Biological Research on Women's Sexual Orientations: Evaluating the Scientific Evidence

Journal of Social Issues, Summer, 2000 by Rosemary C. Veniegas, Terri D. Conley

Summary. The prenatal hormonal theory of women's sexual orientation predicts that women exposed prenatally to excess levels of sex hormones will have masculinized brains and show increased homosexuality. It is unknown whether prenatal exposure to abnormal levels of testosterone actually masculinizes women's brains because the necessary studies have not yet been conducted. Research on CAH indicates that prenatal exposure to excess androgens is weakly associated with some degree of homosexuality in fantasy only among some women. Research on DES indicates that exposure to excess estrogens is modestly associated with greater homosexuality in fantasy and behavior in some women.

Nevertheless, prenatal hormonal exposure falls short of providing a complete explanation for women's sexual orientation because most women have never been exposed. The majority of women exposed to excess androgens or estrogens ultimately develop a heterosexual sexual orientation. Future research on the predicted association between abnormal prenatal hormonal exposure and sexual orientation can be improved by examining lesbians' brains, including appropriate control groups and larger samples, and employing standardized measures of sexual orientation.

The Genetics of Women's Sexual Orientation

Other researchers have investigated the relationship between genes and women's sexual orientation. It has been suggested that homosexuality "runs in families," meaning that there may be a genetic underpinning to the development of sexual orientation. A growing number of studies have examined this issue by assessing the incidence of homosexuality in families, the similarity of sexual orientation between twins, and possible genetic markers for homosexuality. This genetic perspective has been examined in several studies of lesbians, their siblings, and their families.

Incidence of homosexuality in families. Do lesbians have more lesbian sisters and lesbian relatives than do heterosexual women? Typically, in studies of this question, individual women (referred to as probands) are recruited and are asked about the number of homosexual or heterosexual siblings or relatives in their families. Three studies found that lesbians have significantly greater numbers of nonheterosexual siblings (sisters or brothers) than do heterosexual women (Bailey & Bell, 1993; Bailey & Benishay, 1993; Pillard, 1990). In the most recent study Pattatucci and Hamer (1995, p. 408) predicted that "a genetically influenced trait ... should appear at elevated rates in second-degree and third-degree lineages as well as within the nuclear family." (Sisters and daughters are first-degree relatives, children and parents are second-degree relatives, and cousins are third-degree relatives.) They recruited 358 women (probands) for a study on "sexuality in women" through homophile organizations and Women's Stud ies programs. Using very strict criteria to categorize women as lesbian, they found that lesbian probands had significantly higher rates of lesbian sisters, daughters, and cousins through a paternal uncle than did heterosexual women probands. The available data on female homosexuality in families are generally consistent with the genetic perspective but do not rule out the alternative possibility of environmental influences. Pattatucci and Hamer (1995, p. 418) suggested that environmental or social hypotheses might also be applicable to current findings: "social and cultural factors within the family influence the likelihood that a woman will recognize or act on same-sex attractions ... having a bisexual or lesbian sister or mother might make it seem more 'acceptable' for a woman to adopt a nonheterosexual identity." They argued that this family environment explanation could not fully account for the increased incidence of female homosexuality among cousins. Thus, evidence that female homosexuality runs in fa milies is inconclusive because both genetic and environmental hypotheses appear equally applicable to these results.